| holmanb | gschanuel: I have never tried what you're talking about, but the cloud-init vmware datasource docs talk about setting the encoding type for userdata/metadata. I don't see that in the commands you linked above. | 02:33 |
|---|---|---|
| holmanb | gschanuel: https://cloudinit.readthedocs.io/en/latest/topics/datasources/vmware.html?highlight=.encoding | 02:34 |
| Orpheus[m] | New to cloud-init here. How can I "validate" a configuration file without booting a new instance for it? | 02:46 |
| Orpheus[m] | found my answer :) | 02:56 |
| Orpheus[m] | `cloud-init devel schema --config-file file` | 02:56 |
| === IchabodCrane is now known as WrathOfAchilles | ||
| juliank | I just filed https://bugs.launchpad.net/cloud-init/+bug/1949371 | 14:45 |
| ubottu | Launchpad bug 1949371 in cloud-init "'Unable to find a system nic' if openstack lists down ports in json config" [Undecided, New] | 14:45 |
| juliank | This is currently breaking most of ppc64el scalingstack cloud | 14:46 |
| juliank | Arguably the cloud might be broken too | 14:46 |
| juliank | I'm trying to hack an image together that does not error out here, but it's hard | 14:46 |
| juliank | As I first need to get a server booted | 14:46 |
| blackboxsw | Thanks juliank: I think we'd need to see what network_data.json is representing to the instance. As the network JSON there will instruct cloud-init what to do with all interfaces. Since OpenStackLocal datasource reads networkdata using the link-local address http://169.254.169.254/openstack/2018-08-27/network_data.json that data should be available to us into the system via console or ssh. | 17:24 |
| blackboxsw | One way we might do that is using write_files: to emit /etc/netplan/50-cloud-init.yaml for only the known NIC you expect to be up and re-run netplan apply via something like this https://pastebin.canonical.com/p/Jx3cQVDgs7/ | 17:25 |
| blackboxsw | then you can ssh into the instance I hope and `cloud-init query ds.network_json` to see what openstack told us | 17:26 |
| blackboxsw | alternative would be to try logging into the console via the novnc url (which I think you said was unavailable on this openstack deployment) openstack console url show <yourinstance>. on my stack I've got the ability to launch a browser to the URL provided to login without SSH | 17:27 |
| blackboxsw | looks like the internet broke CI again. I see it on brett's and my failing PRs https://github.com/canonical/cloud-init/pull/1090/checks?check_run_id=4070158851 | 17:48 |
| ubottu | Pull 1090 in canonical/cloud-init "Add LXD to datasource list" [Open] | 17:48 |
| holmanb | @blackboxsw - it looks like at least one of the failures can be fixed by cherry picking a test dependency from main | 18:27 |
| holmanb | scratch that, I didn't test well enough locally | 18:31 |
| blackboxsw | looks to me like it may have been a spurious error my next pull is running fine https://app.travis-ci.com/github/canonical/cloud-init/builds/240972200 | 20:09 |
| blackboxsw | falcojr: lxd_container integration test and lxd_vm tests have passed on my PR | 20:14 |
| falcojr | blackboxsw: https://github.com/canonical/cloud-init/pull/1040 merged! | 20:43 |
| ubottu | Pull 1040 in canonical/cloud-init "Add LXD datasource" [Merged] | 20:43 |
| blackboxsw | woo hoo! coffee break. | 20:43 |
| falcojr | blackboxsw: upstream release PR when ready https://github.com/canonical/cloud-init/pull/1091 | 21:24 |
| ubottu | Pull 1091 in canonical/cloud-init "Release 21.4" [Open] | 21:24 |
| blackboxsw | falcojr: grabbing it now | 22:46 |
| blackboxsw | falcojr: approved the PR, but integration tests have yet to run on 21.4 yet now that LXD datasource is committed. I want to get a +1 from the daily integration runs @ https://jenkins.ubuntu.com/server/job/cloud-init-integration-lxd_container-impish-daily/ and https://jenkins.ubuntu.com/server/job/cloud-init-integration-lxd_vm-impish-daily/ before we sign and push the upstream version | 23:12 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!